• Journal of Korean Society of Transportation
• /
• v.24 no.2 s.88
• /
• pp.79-89
• /
• 2006

The purpose of this study is to propose a new method using the GPS-INS integrated system which measures the superelevation of a curved yoad rapidly and safely Also, using the CPS-INS integrated system. this Paper is to analyze the expected accuracy of the superelevation on the roads which are already opened for traffic. The superelevation of the roads which are already opened for traffic often changes by roads rehabilitation work such as resurfacing of Pavement and rutting However, it is difficult to measure 1,he changes of road geometry because or the high cost and the safety problem. If the measurement of superelevation using GPS-INS integrated system is used instead of existing conventional surveying method, the superelevation of road is expected to be measured accurately(under 0.4% RMSE) just by driving without interfering the traffic Therefore, the application of the GPS-INS integrated system is expected to be useful tool for measuring the superelevation of the roads.

• Journal of Korean Society of Transportation
• /
• v.27 no.5
• /
• pp.63-74
• /
• 2009

One of the effective countermeasures for preventing traffic accidents is to provide traffic safety warning information to drivers. Provision of warning information would lead to safer driving to avoid accident occurrence. This study investigated the effects of in-vehicle warning information on driver's behavior. A variety of warning information contents using text, sound, and pictograms were prepared for the field experiments. Individual vehicle speed and acceleration data, which represent quantitative drivers' behavior in response to in-vehicle warning information, were collected using differential global positioning systems (DGPS). Statistical analyses including ANOVA and Tukey's pairwise comparison were conducted. It is expected that the results could be invaluable for designing more effective warning information.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2023.05a
• /
• pp.76-76
• /
• 2023

최근 자율주행과 관련한 시장의 관심은 기존 자동차 자율주행에서 선박 자율운항으로 자연스럽게 이동하고 있다. 이에 인공지능 및 빅데이터 등과 같은 최근 기술을 선박 자율주행에 적용하는 자율운항선박(MASS: Maritime Autonomous Surface Ship) 개발이 활발히 진행되고 있으며, 레이더 및 카메라 등과 같은 센서 정보를 선박 자율운항에 적용하여 다양한 선박 운동 및 정보를 획득하는 연구 기술이 집중되고 있다. 이러한 경향에 따라 IMO(International Maritime Organization)과 같은 국제기구에서도 자율운항선박 표준화 본격 논의로 기술표준 선점 경쟁에 참여하고 있다. 이 중 연안 자율운항선박 개발은 IMO에서 주관하는 무인화 핵심기술로 여겨지고 있어, 기존 대양 항해 기술과 함께 연안 항해에 대한 기술 개발의 중요성이 높아지고 있다. 특히 항만 인근 해역에서는 다수의 선박이 입출항함으로 인해 해상에서의 안전과 물류의 효율화가 요구되기 때문에 고도화된 자율운항 기술개발이 필요하다. 하지만 자율운항선박에서의 상황인식 기술은 탑재된 센서의 제한된 시야각 및 기상조건에 따른 인식률이 떨어지는 문제가 생긴다. 이러한 기술적 한계를 극복하기 위해 육상에 설치된 레이더를 활용하여 선박을 탐지할 수 있는 기술이 필요하다. 본 연구에서는 고해상도 육상 레이더를 기반하여 얻어진 레이더 화면상의 물표 정보를 이용해 인공지능 기법에 활용하기 위한 라벨링 자동 생성 방법에 대해 소개한다. 얻어진 물표 정보에 인공지능 기법을 적용하여 선박 길이 정보를 추정하는 기술에 대해 소개한다.

• Journal of Korean Society of Transportation
• /
• v.33 no.4
• /
• pp.323-336
• /
• 2015

Estimating the risks on the roadway using surrogate safety measures (SSM) has an advantage in that it focuses on the vehicle trajectory directly involved in conflicts. On the other hand, there is a restriction on estimating the risks of continuous segments due to the limited data collected from a location. To overcome the restriction, this study presents the scheme of acquiring the vehicular trajectory using real time kinematics-differential global positioning system (RTK-DGPS) and develops a methodology which contains the considerations of the problems to calculate the SSM such as time-to-collision (TTC), deceleration rate to avoid collision (DRAC) and acceleration noise (AN). By using the methodology, this study shows a result from an experiment executed in a section where the variation of vehicular movement can be observed from several continuous flow roadway sections near Seoul and Gyeonggi Province in Korea. The result illustrated the risks on the roadway by the SSM metrics in certain situations like merging and diverging, stop-and-go, and weaving. This study would be applied to relate the dangers with characteristics of drivers and roadway sections, and prevenst accidents or conflicts by detecting dangerous roadway sections and drivers' behaviors. This study contributes to improving roadway safety and reducing car-accidents.

• Journal of Power System Engineering
• /
• v.16 no.1
• /
• pp.84-90
• /
• 2012

In unmanned vehicles' navigation, the shapes of obstacles are generally irregular and complex. The motion of vehicles based on the range sensor system such as ultrasonic sensors or laser sensors can be unstable due to the irregular shape of the obstacles. In this case, to generate stable trajectory of unmanned vehicles equipped with range sensors, we need an approach that can simplify an obstacle's irregular shape information. In this paper, we propose the trajectory generation algorithm that an vehicle can stably navigate in the environments where irregular shaped obstacles are scattered. The proposed method is verified through the analysis of vehicle's trail and direction data acquired by simulations and implementations.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.1
• /
• pp.105-110
• /
• 2009

In this paper, we proposed a navigation algorithm of the mobile robot for obstacle avoidance using a circular path planning method. The proposed method makes circular paths in order to avoid obstacles in the front side of the mobile robot. An optimal path for approaching to the target is selected and the linear and angular speeds for stable moving of the mobile robot are controlled. Obstacles are detected by image processing which reduce image data obtained from a web camera. Performance of the proposed algorithm is shown by experiments with application to the Pioneer-2DX mobile robot.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.22 no.3
• /
• pp.241-249
• /
• 2014

This paper presents a path planning algorithm of an autonomous vehicle (ADAM III) for collision avoidance in the presence of multiple obstacles. Under the requirements that a low-cost GPS is used and its computation should be completed with a sampling time of sub-second, heading angle estimation is proposed to improve performance degradation of its measurement and a hierarchical structure for path planning is used. Once it is decided that obstacle avoidance is necessary, the path planning consists in three steps: waypoint generation, trajectory candidate generation, and trajectory selection. While the waypoints and the corresponding trajectory candidates are generated based on position of obstacles, the final desired trajectory is determined with considerations of kinematic constraints as well as an optimal condition in a term of lateral deviation. Finally the proposed algorithm was validated experimentally through field tests and its demonstration was performed in Autonomous Vehicle Competition (AVC) 2013.

• Journal of the Korean Society of Industry Convergence
• /
• v.19 no.2
• /
• pp.75-80
• /
• 2016

This study proposes a efficient technology to control the optimal trajectory planning and real-time implementation method which can perform autonomous travelling for unmaned factory automation. Online path planning should plan and execute alternately in a short time, and hence it enables the robot avoid unknown dynamic obstacles which suddenly appear on robot's path. Based on Route planning and control algorithm, we suggested representation of edge cost, heuristic function, and priority queue management, to make a modified Route planning algorithm. Performance of the proposed algorithm is verified by simulation test.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2005.04a
• /
• pp.237-240
• /
• 2005

본 본문은 Chus's 방정식과 Lorenz 방정식을 이용한 카오스 회로를 이동 로봇에 내장하여 카오스 이동 로봇을 구성하고 이 카오스 이동 로봇이 어느 임의 평면을 카오스 궤적을 가지고 주행 또는 탐색하다가 상호 연동 제어 명령이 내려지면 2개 또는 여러 개의 로봇이 서로 연동되어 동기화 되는 방법을 제시하고 그 결과를 검증하였다. 카오스 로봇 동기화에서는 장애물이 있는 경우와 은닉 장애물이 있는 경우에서 2개의 로봇이 서로 연동되어 움직이는지를 확인하였으며 만족할 만한 결과를 얻었다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.11a
• /
• pp.117-120
• /
• 2008

동영상의 다중 물체 인식 및 추적은 의료영상이나 무인 주행 시스템 등의 응용분야에서 중요성이 높아지고 있다. 본 논문에서는 스포츠 동영상의 다중 물체를 인식 및 추적하기 위해 칼만필터 알고리즘을 사용한다. 칼만필터 알고리즘을 이용한 물체의 이동 궤적 관리를 통해 표적 겹침 현상에 대한 추적 실패를 극복하도록 하였다. 표적 겹침이 일어나는 동영상을 입력 영상으로 이용하여 제안한 실시간 시뮬레이터의 추적 성능을 분석하였다.